Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{207102,
        author = {Himanshu Shrivastav and Vishal and Kanishka Bhardwaj and Akhil Pandey},
        title = {An Explainable Machine Learning Framework for Loan Approval Prediction Using Imbalanced Financial Data},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {41-46},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=207102},
        abstract = {Loan approval prediction is a critical task in modern financial systems, where institutions must balance risk minimization with efficient and fair decision-making. Traditional rule-based approaches often fail to capture complex, nonlinear relationships within financial datasets and may lead to inconsistent and biased outcomes. To address these limitations, this study proposes an explainable machine learning framework for loan approval prediction using structured financial data. The proposed framework integrates comprehensive preprocessing techniques, including missing value imputation, categorical feature encoding, feature scaling, and class imbalance handling through the Synthetic Minority Oversampling Technique (SMOTE). Multiple machine learning models—Logistic Regression, Random Forest, Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost)—are implemented and systematically evaluated using key performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. The experimental results demonstrate that ensemble-based models, particularly XGBoost, outperform traditional approaches in terms of predictive accuracy and robustness.In addition to performance enhancement, the framework incorporates explainable artificial intelligence using SHapley Additive Explanations (SHAP), which provides detailed insights into model predictions and identifies the most influential features, including credit history, applicant income, and loan amount. This enhances transparency, supports regulatory compliance, and increases trust in automated decision-making systems. The proposed framework not only improves prediction accuracy but also ensures interpretability and fairness, making it suitable for real-world financial applications. The study contributes to the development of reliable, transparent, and efficient loan approval systems in the era of intelligent financial technologies.},
        keywords = {Credit Risk Assessment, Explainable Artificial Intelligence (XAI), Machine Learning, SHAP, SMOTE, XGBoost.},
        month = {July},
        }